AUTHOR(S): Jeret, Joseph S., M.D.;
Mandell, Menachem, M.D.; Anziska, Brian, M.D.;
Lipitz, Mark, D.O.; Vilceus, Antenor P., M.D.;
Ware, James A., M.D.; Zesiewicz, Theresa A., M.D.
Departments of Neurology (JSJ, BA, ML, APV, JAW,
TAZ) and Radiology (MM), SUNY Health Science
Center at Brooklyn, Brooklyn, New York
Neurosurgery 32; 9-16, 1993
ABSTRACT: WE PROSPECTIVELY STUDIED 712
consecutive patients during a 1-year period who
presented with amnesia or loss of consciousness after
nonpenetrating head trauma and who had a perfect
Glasgow Coma Scale score of 15. Of the 67 (9.4%)
patients with acute traumatic lesions disclosed by
computed tomography (CT) of the head, 2 required
neurosurgical intervention and 1 died. Four factors
were statistically correlated (P < 0.05) with abnormal
CT findings: Older age, white race, signs of basilar
skull fracture, and being either a pedestrian hit by a
motor vehicle or a victim of an assault. Sex, length of
antero- or retrograde amnesia, forward and reverse
digit spans, object recall, focal abnormality on the
general neurological exam, and subjective complaints
were not statistically correlated with CT abnormality.
Using step-wise discriminant function analysis, no
single item or combination of items could be used to
classify 95% of the patients into either the normal or
abnormal CT group. Therefore, regardless of age,
mechanism of injury, or clinical findings, intracranial
lesions cannot be completely excluded clinically on
head-trauma patients who have loss of consciousness
or amnesia, even if the Glasgow Coma Scale score is
15. However, only two patients (0.3%) required
neurosurgical intervention.
KEY WORDS: Coma; Computed tomography;
Glasgow Coma Scale; Head trauma;
Intracerebral hematoma; Mental status examination
The development of the Glasgow Coma Scale (GCS)
(41,42)
and the introduction of computed tomography
(CT) have improved the management of head injury.
However, criteria for emergency neuroradiological
evaluation and hospital admission, decided on the
basis of potential costs and benefits, have not been
established. We designed this study to identify a set
of clinical criteria capable of discriminating those
head-trauma victims with a GCS score of 15 who are
at a particular risk for having significant
abnormalities disclosed by CT.
Previous studies addressing these issues were not
prospective, did not specify details of the mental
status examination, or combined adults and children
into a single study (2,5,7,9,11,16,18,24-26,31-36,39,43,45). We
have attempted to eliminate these disadvantages by

prospectively studying all adult (18 years and older)
patients with loss of consciousness (LOC) or amnesia
and with a GCS score of 15 who arrived in the
emergency room within 24 hours of a head injury
during a 1-year period. Because patients with a GCS
score of 14 or less will almost invariably undergo CT,
they pose less of a decision-making dilemma to the
emergency room physician and were not the focus of
this study.
METHODS
From August 1, 1990, to July 30, 1991, we
examined all patients with blunt head trauma and
LOC or amnesia, regardless of duration, who arrived
at Kings County Hospital, an inner city medical
center and the major trauma center in the borough of
Brooklyn in New York City. Patients were assigned a
GCS score by the consultant, and only patients with a
GCS score of 15 were included in the present report.
Patients transferred from other hospitals, under 18
years of age, with no amnesia or LOC, with
penetrating cranial trauma, or presenting more than
24 hours after the event were excluded.
We designed a checklist to evaluate each patient,
following examples in the literature (20). One of 12
neurologists participating in the study saw the patient
in the emergency department and recorded the GCS
score, age, race, sex, and mechanism of injury.
Patients were asked about the presence of nausea,
vomiting, headache, diplopia, blurred vision, vertigo
or dizziness, tinnitus, and seizure and were
questioned to ascertain the length of retrograde and
posttraumatic amnesia. When available, witnesses
were asked to verify and to supplement these data.
Each patient had a complete neurological
examination. Learning was tested by the ability to
recall the same four objects at 5 minutes. Short-term
memory was assessed by forward digit span recall
and concentration by reverse digit span recall (40).
Of the 753 patients who fulfilled our entry criteria,
41 patients refused CT and were excluded from the
study. The remaining 712 patients underwent
noncontrast CT scans performed on a thirdgeneration CT scanner (either the GE 9800 [General
Electric Medical Systems, Milwaukee, WI] or
Siemens Somatom DR [Siemens Medical Systems,
Inc., Iselin, NJ]). All studies were photographed,
using brain (L, 40; W, 80), bone (L, 250; W, 2000),
and subdural (L, 40; W, 150) windows, and were read
by the neuroradiologist (MM), who did not know the
details of the trauma, the patient's clinical condition,
and the official institutional reading. Basilar skull
fracture was diagnosed by established criteria (3). All
data were entered onto the IBM-4381 (IBM, New
York, NY) and analyzed using SPSSX (SPSS Inc.,
Chicago, IL) by a statistician.
RESULTS
Demographic features
Of the 712 patients who fulfilled our entry criteria,
520 (73%) were male and 192 (27%) were female
(Table 1). Ages ranged from 18 to 90, with a mean ±
SD of 35.6 ± 14.1 years. There were 152 (21%)
whites, 463 (65%) blacks, 69 (10%) Hispanics, and

Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.

Neurosurgery 1992-98
January 1993, Volume 32, Number 1
9 Clinical Predictors of Abnormality Disclosed by
Computed Tomography after Mild Head Trauma
Clinical Study

Computed tomographic findings
There were 67 patients (9.4%) with significant
abnormalities demonstrated by CT (Table 2). Linear
skull fracture and soft tissue swelling in the absence
of other findings were not included in this category.
Thirty-two of the 67 patients (48%) with abnormal
CT scans had multiple abnormalities.
Factors correlated with abnormalities
demonstrated by computed tomography
Four factors were statistically correlated with
abnormal CT scans: Older age, white race, signs of
basilar skull fracture, and certain mechanisms of
injury. The mean age of patients with a normal CT
scan, 34.8 ± 13.5 years, was significantly lower than
those with an abnormal CT scan, 43.1 ± 17.0 years (P
< 0.0005, two-tailed t-test). Whites also had a
significantly higher incidence of abnormalities
disclosed by CT (P < 0.02, two-tailed t-test). Of the
25 patients with clinical evidence of basilar skull
fracture, 8 had an abnormal CT scan (P < 0.0001, ttest with Pearson correction). CT demonstrated
abnormality was also more common in cases of
assault and for pedestrians struck by motor vehicles
(P < 0.02, two-tailed t-test).
Factors not correlated with abnormal computed
tomographic scans
Subjective complaints after acute head trauma were
very common (Table 3). Most patients complained of
headache. However, none of the subjective
complaints studied was statistically correlated with
the presence of abnormal CT scans (P > 0.05 for
each, two-tailed t-test). Length of retrograde and posttraumatic amnesia were also not statistically
correlated with abnormal CT findings (P = 0.8 and
0.40, two-tailed t-test, respectively). Focal
abnormalities on the general neurological
examination were present in 26 patients, 22 of whom
had a normal CT scan (P > 0.05).
The mean forward and reverse digit span and
object recall for patients with normal CT scans (6.1 ±
1.5, 3.6 ± 1.5, and 2.1 ± 1.3, respectively) was not
significantly greater than for those with abnormal CT
scans (6.1 ± 1.8, 3.6 ± 1.6, and 2.1 ± 1.4,
respectively), P > 0.05 for each (Table 4). In some
cases, performance on our mental status testing was
entirely normal, including some of the patients with
the most significant intracranial lesions disclosed by
CT, i.e., those with intracranial hematomas,
intraventricular hemorrhage, and depressed skull
fracture (Table 5). Of note, only three patients had a
very significant adverse outcome (one death, two
requiring craniotomy). Using step-wise discriminant
function analysis, no single item or combination of
items could be used to classify 95% of patients into
either the normal or abnormal CT group.
Significant case reports

Patient 71
A 49-year-old man was found beside his bicycle
with his wallet missing and significant external
bruises including a large hematoma over the right
eye. At arrival at the emergency department, he was
alert, oriented, and able to do simple calculations. His
forward and reverse digit span tests were 7 and 4,
respectively. His general neurological examination
was normal; specifically, there was no facial
weakness. A fracture of the right clavicle and several
rib fractures were diagnosed. A peritoneal lavage
disclosed nothing abnormal. During the next several
hours, he became more lethargic, agitated, and
confused; intravenous administration of lorazepam
was necessary. A CT scan revealed a left temporal
contusion and right-side subarachnoid hemorrhage.
Six hours after arrival, a mild left central facial
paresis was noted. Craniotomy was performed that
day for debridement of the contusion. The patient did
well postoperatively. Neuropsychological testing
demonstrated comprehension deficits, paraphasic
errors, and dysnomia--all consistent with Wernicke
aphasia. He was discharged to his home on hospital
Day 24.
Patient 297
A 41-year-old alcoholic man was assaulted late one
evening but went home without seeking medical
attention. When he awakened several hours later with
headache, he was brought to the hospital. He had
several minutes of amnesia and complained of
headache, nausea, three episodes of vomiting, and
slight dizziness. Despite an alcohol level of 108
mg/dL, he was alert and oriented. His forward and
reverse digit span tests were 4 and 3, respectively.
His general neurological examination was normal.
Head CT scan demonstrated an acute right subdural
hematoma. He underwent successful craniotomy and
removal of the hematoma and was discharged to his
home on hospital Day 14.
Patient 530
A 79-year-old, unrestrained man fell asleep while
driving a car and struck a pole. At his arrival in the
emergency room, he was awake, oriented, and
complaining only of mild headache. His medical
history was significant for diabetes mellitus,
hypertension, congestive heart failure, chronic renal
insufficiency, and previous myocardial infarction. He
was able to repeat 7 digits forward and 3 digits
backward. Except for presbycusis and bilateral
periorbital ecchymosis, his examination was normal.
X-rays failed to reveal any fractures. Head CT scan
demonstrated an intracranial air-fluid level anterior to
the left frontal lobe, a left subdural hematoma,
pneumocephalus, and cerebral atrophy. He began
receiving intravenous antibiotics. A repeat CT scan
on hospital Day 4 showed resolution of the
intracranial lesions. On hospital Day 5, he became
hypotensive, bradycardic, and lethargic; he was
intubated. Cardiac isoenzymes were consistent with
myocardial infarction. He developed recurrent
episodes of sustained ventricular tachycardia that
were unresponsive to medical therapy and pacemaker

Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.

28 (4%) members of other racial groups. Traffic
accidents (n = 400) and assaults (n = 218) were the
most common causes of injury.

DISCUSSION
Indications for neuroimaging in acute trauma have
not been established. Different study populations
among previous reports make generalization difficult
(Table 6). Some have recommended that all patients
undergo CT, even in the absence of LOC and
neurological deficit (18), or if there is progressive
headache, vomiting, serious facial injury, or multiple
trauma (23). Others take the mechanism of injury into
account as well (22,38). Some require LOC, amnesia,
signs of possible depressed skull fracture, or focal
neurological deficits (24,28,39).
Because 5% of patients with a GCS score of 13 to
15 may eventually require neurosurgery (39), we
searched for other clinical and epidemiological
factors that may provide a more sensitive and specific
predictor of CT abnormality. Patients with a GCS
score of 14 or less should ideally undergo CT
whenever possible. For those with a GCS score of 15,
clinical management is less clear and has not been
adequately studied. One report included 89 patients
with a GCS score of 15 who had extradural
hematomas (37). Despite analysis of historical and
epidemiological factors, subjective complaints, and
mental status examination, we found no item or
combination of items that could alter the treatment
algorithm for head trauma patients with a GCS score
of 15 and that would provide an accurate prediction
of CT normality or abnormality.
Of the epidemiological descriptors studied, age
was most significantly (P < 0.0005) associated with
an abnormal CT scan. Increased mortality and
morbidity with increasing age is well documented (1,
6,13,15,30)
and confirmed by our study. Although men
may be 3 to 4 times more likely to suffer lethal injury
(8,44)
, we did not observe any difference in the
incidence of CT abnormality between the sexes. As in
previous studies of head trauma in the United States,
traffic accidents accounted for about half the cases (6,
12,14,17,32,33,45)
. Victims of assault and pedestrians
struck by a motor vehicle had an abnormal CT more
often. One report found that being hit with a blunt
object (e.g., baseball bat) was less often associated
with intracranial abnormality than being assaulted
with fists and feet (10). This is not confirmed by our
study where assault with a blunt object was twice as
likely to cause CT abnormality (17% versus 9%). The
higher incidence of CT disclosed abnormality in
white patients (P = 0.03, two-tailed t-test) likely
represents selection bias; less severely injured white
patients may have sought care at a site other than a
city hospital. Other studies have found the incidence
of head trauma in major cities to be highest for
minority populations (28,37).
A recent British editorial recommended hospital
admission for all patients with a GCS score of less
than 15, signs of basilar skull fracture, seizure, focal
neurological signs, persistent headache or vomiting,
and skull fracture (27). Because CT is more readily
available in the United States, patients with linear
skull fractures, a GCS score of 15, and no intracranial

abnormalities disclosed by CT can be safely
discharged with appropriate instructions and followup (21). Admission and observation is recommended
for all patients with intracranial hematomas. The high
frequency of intracranial bleeding in our study
population of patients with mild head trauma and a
GCS score of 15 has not been previously
documented. Nonetheless, only three had a very
significant adverse outcome. However, Patient 71
would have undergone CT anyway because his
mental status deteriorated and he developed a left
facial paresis; Patient 297 would have undergone CT
because of progressive headache, especially because
he was alcoholic and at higher risk for intracranial
hematoma; and Patient 530 died of multiple medical
complications despite radiological evidence of
resolution of his intracranial lesions.
If the clinician has a heightened level of suspicion,
it is appropriate to perform a head CT scan even if the
patient is fully oriented, has a GCS score of 15, and
normal performance on bedside neuropsychological
testing. Criteria for CT in patients without LOC or
amnesia remain to be determined by a prospective
study.
Although several facets of the neurological and
mental status examinations were statistically
correlated with CT abnormality in our study, a
normal clinical and mental status examination neither
excluded significant head injury nor eliminated the
role of neuroimaging. Patients with LOC or amnesia
after head trauma may harbor intracranial lesions,
even if they have a GCS score of 15. However, the
chance of requiring neurosurgical intervention is
small.
ACKNOWLEDGMENTS
The authors wish to thank Matthew J. Avitable,
Ph.D. and Richard Lechtenberg, M.D. for performing
the statistical analyses of our data. We also
acknowledge with appreciation the helpful
suggestions and contributions of Drs. Roger Q.
Cracco, Mahendra Somasundaram, Paul Cooper,
Jeffrey Benjamin, Ronald Lazar, Gwendolyn Hotson,
Luis Figueroa, Katie Henry, Carrie Landess, Antoin
Munirji, Moquit Usman, Michael Goodman, and
Aristotles Pena. We are also indebted to research
assistants Melanie Sinatra and Ian Cole. The
cooperation of the residents in the departments of
neurology, neurosurgery, radiology, and trauma
surgery is also appreciated. Financial assistance for
the research assistants and for expenses related to
presenting the article were provided by New Medico
(Lynn, MA) and SUNY Health Science Center at
Brooklyn (Brooklyn, NY).
Presented in part at the 41st Annual Meeting of the
Congress of Neurological Surgeons, Orlando, Florida,
1991.
Received, February 11, 1992.
Accepted, July 7, 1992.
Reprint requests: Joseph S. Jeret, M.D., 3126
Kings Highway, Suite 2, Brooklyn, NY 11234.
REFERENCES: (1-45)

Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.

overdrive. On hospital Day 8, he became hypotensive
and died.

2.

3.

4.

5.
6.

7.

8.

9.

10.

11.

12.

13.
14.

Alves WM, Jane JA: Mild brain injury:
Damage and outcome, in Becker DP,
Povlishock JT (eds): Central Nervous System
Trauma Status Report. Rockville, MD,
National Institutes of Health, 1985.
Berger MS, Pitts LH, Lovely M, Edwards
MSB, Bartkowski HM: Outcome from severe
head injury in children and adolescents. J
Neurosurg 62:194-199, 1985.
Cohen WA, Kricheff II: Craniocerebral
trauma, in McCort JJ, Midelzun RE (eds):
Trauma Radiology. New York, ChurchillLivingstone, 1990.
Dacey RG, Alves WM, Rimel RW, Winn R,
Jane JA: Neurosurgical complications after
apparently minor head injury. J Neurosurg
65:203-210, 1986.
Dublin AB, French BN, Rennick JM:
Computed tomography in head trauma.
Radiology 122:365-369, 1977.
Eisenberg HM: Outcome after head injury:
General considerations and neurobehavioral
recovery. Part I: General considerations, in
Becker DP, Povlishock JT (eds): Central
Nervous System Trauma Status Report.
Rockville, MD, National Institutes of Health,
1985.
Feuerman T, Wackym PA, Gade GF, Becker
DP: Value of skull radiography, head
computed tomographic scanning, and
admission in cases of minor head injury.
Neurosurgery 22:449-453, 1988.
Frankowski RF, Annegers JF, Whitman S:
Epidemiological and descriptive studies. Part
I: The descriptive epidemiology of head
trauma in the United States, in Becker DP,
Povlishock JT (eds): Central Nervous System
Trauma Status Report. Rockville, MD,
National Institutes of Health, 1985.
French BN, Dublin AB: The value of
computerized tomography in the management
of 1000 consecutive head injuries. Surg
Neurol 7:171-183, 1977.
Guyot AM, Alcantara A, Shah AK, Piotrowski
C, Roszler MH, Ham SD, Diaz FG, Peterson
PL: Traumatic brain injury due to assault: "Hit
me with a bat please." Proceedings of the
American Academy of Neurology, in
Neurology 41(Suppl 1):308, 1991 (abstr).
Han JS, Kaufman B, Alfidi RJ, Yeung HN,
Benson JE, Haaga JR, El Yousef SJ, Clampitt
ME, Bonstelle CT, Huss R: Head trauma
evaluated by magnetic resonance and
computed tomography: A comparison.
Radiology 150:71-77, 1984.
Jagger J, Levine JI, Jane JA, Rimel RW:
Epidemiological features of head injury in a
predominantly rural population. J Trauma
24:40-44, 1984.
Jennett B, Teasdale G: Management of Head
Injuries. Philadelphia, FA Davis Co., 1981.
Kalsbeek WD, McLaurin RL, Harris BSH,
Miller JD: The national head and spinal cord

15.

16.
17.

18.
19.

20.
21.

22.
23.

24.
25.

26.

27.
28.

29.

injury survey: Major findings. J Neurosurg
53(Suppl):19-31, 1980.
Klauber MR, Marshall LF, Luerssen TG,
Frankowski R, Tahaddor K, Eisenberg HM:
Determinants of head injury mortality:
Importance of the low risk patient.
Neurosurgery 24:31-36, 1989.
Koo AH, LaRoque RL: Evaluation of head
trauma by computed tomography. Radiology
123:345-350, 1977.
Kraus JF, Black MA, Hessol N, Ley P, Rokaw
W, Sullivan C, Bowers S, Knowlton S,
Marshall L: The incidence of acute brain
injury and serious impairment in a defined
population. Am J Epidemiol 119:186-201,
1984.
Leclercq TA, Bolton R: Indications for CT
scanning in benign head trauma. RI Med J
68:503-505, 1985.
Levander B, Stattin S, Svendsen P: Computer
tomography of traumatic intra- and
extracerebral lesions. Acta Radiol [Suppl]
(Stockh) 346:107-118, 1975.
Lindgren S: Acute head injury management
charts. Acta Neurochir [Suppl] (Wien) 36:95102, 1986.
Livingston DH, Loder PA, Koziol JM, Hunt
CD: The use of CT to triage patients requiring
admission following minimal head injury.
Proceedings of the American Association for
the Surgery of Trauma, in J Trauma 30:924,
1990 (abstr).
Marshall LF: Head Injury. Presented at the
39th Annual Meeting of the American
Academy of Neurology, New York, 1987.
Masters SJ, McClea PM, Arcarese JS, Brown
RF, Campbell JA, Freed HA, Hess GH, Hoff
JT, Kobrine A, Koziol DF, Marasco JA,
Merten DF, Metcalf H, Morrison JL, Rachlin
JA, Shaver JW, Thornbury JR: Skull x-ray
examinations after head trauma.
Recommendations by a multidisciplinary
panel and validation study. N Engl J Med
316:84-91, 1987.
McMicken DB: Emergency CT head scans in
traumatic and atraumatic conditions. Ann
Emerg Med 15:274-279, 1986.
Mendelow AD, Campbell DA, Jeffrey RR,
Miller JD, Hesset C, Bryden J, Jennett B:
Admission after mild head injury: Benefits
and costs. Br Med J [Clin Res] 285:15301532, 1982.
Mendelow AD, Teasdale G, Jennett B, Bryden
J, Hessett C, Murray G: Risks of intracranial
haematoma in head injured adults. Br Med J
[Clin Res] 287:1173-1176, 1983.
Miller JD: Assessing patients with head
injury. Br J Surg 77:241-242, 1990.
Miller JD, Murray LS, Teasdale GM:
Development of a traumatic intracranial
hematoma after a "minor" head injury.
Neurosurgery 27:669-673, 1990.
Paxton R, Ambrose J: The EMI scanner. A
brief review of the first 650 patients. Br J

Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.

1.

31.

32.
33.

34.

35.

36.
37.

38.

39.

40.
41.
42.
43.

44.

45.

Radiol 47:530-565, 1974.
Pentland B, Jones PA, Roy CW, Miller JD:
Head injury in the elderly. Age Ageing
15:193-202, 1986.
Plaut MR, Gifford RRM: Trivial head trauma
and its consequences in a perspective of
regional health care. Milit Med 141:244-247,
1976.
Rimel RW, Giordani B, Barth JT, Jane JA:
Disability caused by minor head injury.
Neurosurgery 9:221-228, 1981.
Rimel RW, Giordani B, Barth JT, Jane JA:
Moderate head injury: Completing the clinical
spectrum of brain trauma. Neurosurgery
11:344-351, 1982.
Rivara F, Tanaguchi D, Parish RA, Stimac
GK, Mueller B: Poor prediction of positive
computed tomographic scans by clinical
criteria in symptomatic pediatric head trauma.
Pediatrics 80:579-584, 1987.
Ros SP, Ros MA: Should patients with normal
cranial CT scans following minor head injury
be hospitalized for observation? Pediatr
Emerg Care 5:216-218, 1989.
Roy CW, Pentland B, Miller JD: The causes
and consequences of minor head injury in the
elderly. Injury 17:220-223, 1986.
Servadei F, Faccani G, Rocella P, Seracchioli
A, Godano V, Ghadirpour R, Naddeo M,
Piazza G, Carrieri P, Taggi F, Pagni CA:
Asymptomatic extradural hematomas. Results
of a multicenter study of 158 cases in minor
head injury. Acta Neurochir (Wien) 96:3945, 1989.
Shah AK, Guyot AM, Ham SD, Diaz FG,
Peterson PL: "CT or not to CT": ER
evaluation of head trauma. Proceedings of the
American Academy of Neurology, in
Neurology 41(Suppl 1):308, 1991 (abstr).
Stein SC, Ross SE: The value of computed
tomographic scans in patients with low-risk
head injuries. Neurosurgery 26:638-640,
1990.
Strub RL, Black FW: The Mental Status
Examination in Neurology. Philadelphia, FA
Davis Co., 1988.
Teasdale G, Jennett B: Assessment of coma
and impaired consciousness. Lancet 2:81-83,
1974.
Teasdale G, Jennett B: Assessment and
prognosis of coma after head injury. Acta
Neurochir (Wien) 34:45-55, 1976.
Wahab A, Ibrahim M, Chowdhary UM,
Ammar A, Ibrahim EM: The outcome of head
injuries: The Saudi experience. Nagoya J
Med Sci 51:1-6, 1989.
Whitman S, Coonley-Hoganson R, Desai BT:
Comparative head trauma experiences in two
socioeconomically different Chicago-area
communities: A population study. Am J
Epidemiol 119:57-80, 1984.
Zimmerman RA, Bilaniuk LT, Gennarelli T,
Bruce D, Dolinskas C, Uzzell B: Cranial
computed tomography in diagnosis and

management of acute head trauma. AJR
131:27-34, 1978.

COMMENTS
The management of mild head injury has been an
area of some controversy because of an absence of
established guidelines. Although the vast majority of
patients with mild head injury go on to make
uneventful recoveries, a few patients deteriorate and
become neurologically devastated or die. The
incidence of such deterioration is very low (i.e., less
than 0.02%) in other reported studies and 0.4% in the
present report. Nevertheless, in the current
medicolegal climate in the United States, there is zero
tolerance for such an eventuality. The only effective
way to avoid missing such cases is to obtain
computed tomographic (CT) scans on every patient
with a head injury. Even in a very highly developed
and wealthy medical system, such as in the United
States, this is rather expensive and not always
practical. It is important therefore to try to identify
subgroups of patients who are at particular risk for
developing intracranial masses and to obtain CT
scans on them. Conversely, patients who do not meet
these criteria could be managed with clinical
observation alone. However, this goal has yet to be
achieved.
This article reports an epidemiological study that is
well designed and effectively presented. The study
demonstrates that of 712 consecutive patients with a
Glasgow Coma Scale of 15 who presented to the
emergency room with amnesia or loss of
consciousness after closed-head injury, 67 (9.4%) had
an abnormality on the initial CT scan. Of these, only
two required neurosurgical intervention and one died.
At the most, this represents a 0.4% incidence of
surgical mass lesions in this group of patients. As the
authors have noted in their discussion, other reports
have suggested that adults with no skull fracture and
preserved orientation did not need CT scans because
the chance of significant intracranial hematomas was
less than 0.02%. Although this study also
demonstrates that it is not easy to predict which
patients will have CT abnormalities, it confirmed the
clinical impression that the vast majority of patients
who meet this description will go on to make an
uneventful recovery. In the final analysis, the issue
rests on the question of why a CT scan is needed. If
its purpose is to detect mass lesions because they
cause the patient to deteriorate, this does not seem to
be a very cost effective or practical way to deal with
this group of patients (yield of only 0.4%). A CT scan
could certainly be justified if the clinician has a
heightened level of suspicion for any reason, or if the
scan is needed for academic or research purposes.
Having been surprised on more than one occasion by
rather significant CT scan findings in the presence of
a relatively benign clinical presentation, I believe that
in the perfect world where there are no practical or
financial constraints, scanning every patient with a
mild head injury could be justified. However, this

Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.

30.

study should not be construed to indicate that routine
CT scanning in this subset of patients is a mandatory
or a high-yield activity.

This well-executed, prospective study addresses
the question of how frequently abnormalities are
found on a head CT scan, and whether these lesions
can be correlated with certain demographic or clinical
data in adult patients who suffered a head injury with
loss of consciousness and/or amnesia, but who had a
normal Glasgow Coma Scale score. In 712
consecutive patients, the incidence of post-traumatic
lesions was 9.4% and these lesions could be
correlated with only a few of the many factors
examined. In fact, it is impossible to predict which
patients will and which will not have CT
abnormalities. Does this now mean that every patient
fitting the entry criteria for this study needs to be
scanned? Absolutely not! This study was not
designed to assess the effect of using CT scans on
outcome in these patients, and it also does not
address costs. The only patient who died did so
despite having had a head CT, and a scan is obviously
not insurance against an unfavorable outcome.
Moreover, of the only two patients (less than 0.3%)
who underwent neurosurgical intervention, one
deteriorated before CT scanning and he would have
been scanned (and undergone surgery) anyway,
whereas the other might well have ended up with a
chronic subdural hematoma weeks later. On top of
this, most neurosurgeons have a patient, or know of
one, with a normal CT scan who still needed to be
taken to the operating room emergently hours or days
later. As the authors noted in the introduction,
"criteria for emergency neuroradiological evaluations
and hospital admission, decided on the basis of
potential costs and benefits, have not been
established." Such criteria can probably only be
established in a randomized study, and considering
the enormous financial implications of our decisions,
based on sheer numbers, such a study might well be a
worthwhile undertaking. Until then, however,
deciding "to scan or not to scan," or "to admit or not
to admit" remains more art than science, with most
physicians probably erring "to stay on the safe side."
J. Paul Muizelaar
Richmond, Virginia

Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.

Raj K. Narayan
Houston, Texas

Table 2. Abnormalities Encountered by Computed
Tomography

Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.

Table 1. Demographic Features

Table 4. Mental Status Testing

Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.

Table 3. Subjective Complaintsa

Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.

Table 5. Features of Patients with Most Significant
Abnormality Demonstrated by Computed
Tomographya

Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.

Table 6. Previous Studies of Abnormality
Demonstrated by Computed Tomography after Mild
Trauma