Cranial Computed Tomography

Interpretation in Acute Stroke

Physician Accuracy in Determining Eligibility
for Thrombolytic Therapy
David L. Schriger, MD, MPH; Mary Kalafut, MD, MS; Sidney Starkman, MD;
Michelle Krueger, MD; Jeffrey L. Saver, MD
Context.Intracranial hemorrhage must be excluded prior to administration of
thrombolytic agents in acute stroke.
Objective.To evaluate physician accuracy in cranial computed tomography
scaninterpretationfor determiningeligibility for thrombolytic therapy inacutestroke.
Design.Administration of randomly selected, randomly ordered series of 15
computed tomography scans from a pool of 54 scans that demonstrated intrace-
rebral hemorrhage, acute infarction, intracerebral calcications (impostor for hem-
orrhage), old cerebral infarction (impostor for acute infarction), and normal findings.
Participants.A convenience sample of 38 emergency physicians, 29 neu-
rologists, and 36 general radiologists.
Main Outcome Measures.Physician determination of eligibility for thrombo-
lytic therapy based on computed tomography scan interpretation.
Results.Average correct score by all physicians on all computed tomography
scans was 77% (95% condence interval, 74%-80%). Of 569 computed tomogra-
phy readings by emergency physicians, 67% were correct; of 435 readings by
neurologists, 83% were correct; and of 540 readings by radiologists, 83% were
correct. Overall sensitivity for detecting hemorrhage was 82%(95%condence in-
terval, 78%-85%); 17%of emergency physicians, 40%of neurologists, and 52%of
radiologists achieved 100% sensitivity for identication of hemorrhage.
Conclusion.Physicians in this study did not uniformly achieve a level of sen-
sitivity for identication of intracerebral hemorrhage sufficient to permit safe selec-
tion of candidates for thrombolytic therapy.
JAMA. 1998;279:1293-1297
From the Emergency Medicine Center (Drs
Schriger, Starkman, and Krueger) and the Depart-
ment of Neurology (Drs Kalafut, Starkman, and
Saver), School of Medicine, and the Stroke Center
(Drs Kalafut, Starkman, and Saver), University of
California at Los Angeles.
Dr Starkman has received contracts and grants as
an investigator in acute stroke studies sponsored by
Genentech Inc, South San Francisco, Calif, and
Boehringer-Ingelheim Pharmaceuticals Inc, Ridge-
field, Conn. He also is a member of the Acute Ischemic
Stroke Advisory Board and the Stroke Speakers Bu-
reau, and he has participated in stroke conferences
sponsored by Genentech Inc and Boehringer-
Ingelheim Pharmaceuticals Inc.
Dr Saver has participated as an investigator in re-
search trials fundedby Genentech Inc andBoehringer-
Ingelheim Pharmaceuticals Inc, and has received
honoraria from Genentech Inc for educational
speeches. He also has served on a scientific advisory
board for Boehringer-Ingelheim Pharmaceuticals Inc.
Roche Laboratories Inc, Nutley, NJ, has a substantial
interest in Genentech Inc. These companies have sup-
ported the activities of the University of California at Los
Angeles Stroke Center with unrestricted educational
grants.
Corresponding author: David L. Schriger, MD, MPH,
924WestwoodBlvd, Suite300, LosAngeles, CA90024-
2924 (e-mail: schriger@ucla.edu).
RECOMBINANT tissue plasminogen
activator has been approved for select
patients within 3 hours of onset of acute
ischemic stroke.
1
Since thrombolytic
therapy may produce lethal bleeding in
patients with intracranial hemorrhage,
the presence of intracranial blood on the
initial computed tomography (CT) scan
has been an exclusion criterion in the
5 trials of thrombolytic therapy for
stroke
2-6
and in the recommendations of
expert panels.
7-9
Thus, before thrombo-
lytic therapy may be given in stroke, a
physicianhighlyskilledinidentifyingin-
tracranial hemorrhage must interpret
the CT scan.
For editorial comment see p 1307.
Early signs of major cerebral infarc-
tion (sulcal effacement, mass effect, and
edema) also are associated with an in-
creased risk for intracerebral hemor-
rhage in patients who receive thrombo-
lytictherapy,
10-12
andsomeguidelinesrec-
ommend avoiding thrombolytic therapy
when these findings are present.
7-9
We conducted this study to determine
howwell emergency physicians, neurolo-
gists, and general radiologists identified
cranial CTscans that have evidence of in-
tracranial hemorrhage. We also assessed
these physicians accuracy in interpret-
ing other CT findings commonly seen in
patients withacuteischemic stroketoex-
amine whether physicians could distin-
guish CT scans that demonstrate subtle
imaging abnormalities representing po-
JAMA, April 22/29, 1998Vol 279, No. 16 Cranial CT InterpretationSchriger et al 1293
1998 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ on 02/25/2013
tential contraindications to thromboly-
tics from those that do not preclude safe
administration of these agents.
METHODS
Development of Scan Library
We reviewed interpretations of all
cranial CT scans performed at a univer-
sity teaching hospital from December
1994 to January 1996 to identify scans
exhibiting hemispheric parenchymal
hemorrhage or early infarction. Scans
withcalcifications (usedas impostors for
hemorrhage), scans with old infarction
(used as impostors for acute infarction),
and normal scans without calcification
also were identified. Scans with other
abnormalities (including subarachnoid
and extracerebral hemorrhage), scans
with multiple findings, and scans that
could not be definitively placed in 1 of
these categories were excluded. Only
scans for which there was unanimous di-
agnostic agreement among the authors
(andaconsultingneuroradiologist) were
included. Using a consensus process, we
classified each scan that demonstrated
hemorrhage as easy or difficult to inter-
pret and each scan that demonstrated
acute infarction as easy, intermediate,
or difficult to interpret based on the
subtlety of the findings.
Subject Recruitment
Thirty-eight emergency physicians
were recruited at the Scientific Assem-
bly of the California Chapter of the
American College of Emergency Physi-
cians in May 1996. Seventeen commu-
nity-basedneurologistswhoperiodically
attend at a university hospital neurol-
ogyclinicweretestedinprivatesessions
and 12 additional neurologists were
testedwhile attendingthe Universityof
California at Los Angeles (UCLA)
Stroke Center symposium, held in Octo-
ber 1996. Seven radiologists were con-
tacted through the physician directors
of local community hospital radiology
departments and tested at their hospi-
tals. Twenty-nine other radiologists
were tested at the 49th Annual Midwin-
terRadiological/OncologyConferenceof
the Los Angeles Radiological Society,
held in January 1997. Board-certified
and board-eligible radiologists who did
not have additional training in neurora-
diology were tested.
At the meetings, a booth was set up in
the exhibits area of the conference with
a sign reading Test Your Skills at CT.
Subjects included physicians who spon-
taneously approached the booth and
those who, when approached, agreed to
participate in the study. At conference
test sites, subjectswereofferedaT-shirt
as an incentive to participate and were
informed that the physician in each spe-
cialty with the highest score would re-
ceiveatextbookof hisorherchoice. Phy-
sician-subjects who gave verbal in-
formedconsent toparticipatecompleted
a single-sheet questionnaire regarding
their age, years of clinical experience,
residency training, board certification,
and typical involvement in the reading
of cranial CTscans. Aftercompletingthe
questionnaire, each subject was shown
the CT scans on a view box in an indi-
vidual session with unlimited time to
interpret each scan. The study was ap-
proved by the UCLA Institutional Re-
view Board.
Scan Presentation
Physician-subjects were asked to as-
sume that each scan was of a patient who
arrivedatthehospital withinthefirstfew
hours after the onset of an acute hemi-
spheric neurologic deficit (eg, aphasia,
hemiparesis). As each CT scan was pre-
sented, the physician-subject was told
which side of the patients body was af-
fected. Subjects were asked to accept
that each patient was eligible for throm-
bolytictherapyprovidedtheCTscanhad
no contraindications. For each patient,
thephysicianwasasked, Basedsolelyon
scan findings, could thrombolytics be
administered to this patient? Answer
choices were (1) yes; (2) no, because of
hemorrhage; or (3) no, because of signs of
acute infarction. A list of contraindica-
tions (hemorrhage, early hypodensity,
mass effect, and shift) and a list of find-
ings that did not preclude the adminis-
trationof thrombolytics (calcification, at-
rophy, and old infarction) were provided
to remind subjects of the criteria for this
study. Subjects were informed that 20%
to 60% of the scans would have no con-
traindication to thrombolytic therapy.
Eachsubject waspresentedwith5ini-
tial scans: 2 difficult hemorrhages, 1 in-
termediate acute infarction, 1 impostor,
and 1 normal (Figure). Subjects who re-
sponded correctly to all 5 scans were
placed in an advanced track and were
then presented with 10 scans: 3 difficult
hemorrhages, 3 difficult acute infarc-
tions, 1 intermediate acute infarction, 2
impostors, and 1 normal. Subjects who
respondedincorrectlyto1or moreof the
first 5 scans were placed in the standard
trackandpresentedwith3difficult hem-
orrhages, 1easyhemorrhage, 1interme-
diate acute infarction, 2 easy acute in-
farctions, 1impostor, and2normal scans.
The2-trackstrategywasdesignedtoen-
sure that subjects were given scans that
wereappropriatetotheir skill levels and
would maximize the discriminative ca-
pacity of the test.
13,14
Testing Protocols
Two hundred protocols, each contain-
ing a script of 25 scans, were created.
A B C D
Easy Hemorrhage,
24 (100%) of 24
Difficult Hemorrhage,
25 (76%) of 33
Easy Acute Infarction,
17 (81%) of 21
Intermediate Acute Infarction,
23 (53%) of 43
Single slices from 4 axial computed tomography (CT) scans. These illustrative slices depict the pathology that the physician needed to recognize. Subjects were
shown the patients entire scan in standard format. Data shown are number (percentage) of correct readings for each scan.
1294 JAMA, April 22/29, 1998Vol 279, No. 16 Cranial CT InterpretationSchriger et al
1998 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ on 02/25/2013
Each protocol specified that a subject
must receive an initial series of 5 scans,
with the scans to be included and their
order of presentation determined using
the uniform randomnumbers function
of STATA 5.0 (Stata Corp, College Sta-
tion, Tex). By a similar method, addi-
tional series of 10 scans each were pre-
pared for the standard and advanced
tracks.
Statistical Methods
We designed the experiment to pro-
videstableestimatesof theperformance
of each subject, the difficulty of each
scan, and the sensitivity for detecting
hemorrhage, without unduly burdening
the volunteers. Using a conservative
simulation, we determined that 30 sub-
jects per specialty, each reading 15
scans, would meet these goals. Ninety-
five percent confidence intervals (CIs)
surrounding percent correct values
were calculated using robust clustered
logistic regression, which accounts for
the fact that the scan readings may be
associated with the skills of the reader
and, therefore, are not completely inde-
pendent.
15
RESULTS
All 29 neurologists, all 36 radiologists,
and 74%of the 38 emergency physicians
wereboard-certified(another13%of the
emergency physicians were senior resi-
dents in emergency medicine). Emer-
gency physicians averaged 9 years in
postresidency practice and 36 clinical
hours per week, neurologists averaged
13 years in practice and 42 clinical hours
per week, and radiologists averaged 15
years in practice and 42 clinical hours
per week. Twenty-four percent of emer-
gency physicians routinely read cranial
CT scans. All neurologists reported
reading CT scans; roughly half of them
did so before seeing the radiologists re-
port. All of the radiologists spent some
time reading cranial CT scans (15% of
clinical practice on average), although
22% reported that the interpretation of
cranial CT and magnetic resonance im-
aging scans constituted less than 5% of
their practice.
There were few violations in protocol.
One emergency physician was inadvert-
entlygivenonly14scans. Oneradiologist
was placed in the standard track despite
achieving a perfect score on the first 5
scans. Also, during testing of the first
group of radiologists it became clear that
2 scans, an intermediate acute infarction
and a difficult hemorrhage, were causing
unintended confusion. The difficult hem-
orrhage hadanarea that couldhave been
readas infarction, andsubjects were ret-
rospectivelygivencredit foracorrect an-
swerprovidedtheydidnot scorethescan
as give thrombolytics. As a result of
this, another radiologist was mistakenly
placed in the standard track. These 2
cases were removed and replaced with
alternate scans that did not have these
ambiguities.
Among all 103 physicians, 80 (78%) in-
correctly interpreted at least 1 of the 5
scans in the initial series (Table 1). Five
emergency physicians, 6 neurologists,
and 10 radiologists correctly responded
to all 5 scans and entered the advanced
track, achieving average total scores of
67%, 80%, and 84%, respectively. Thus,
even the best performers misinter-
preted a substantial number of scans.
Across all examinations, neurologists
and radiologists were 100%accurate for
identifying easy hemorrhages. Emer-
gencyphysiciansidentified94%(95%CI,
84%-98%) of these(Table2). Fordifficult
hemorrhages, emergency physicians
correctly read 56% (95% CI, 46%-67%),
neurologistscorrectlyread78%(95%CI,
68%-85%), and radiologists correctly
read 80% (95% CI, 71%-87%), for an
overall sensitivity for hemorrhage of
82% (95% CI, 78%-85%). Fifty-two per-
cent of radiologists, 40%of neurologists,
and 17% of emergency physicians cor-
rectly identified all CT scans with evi-
dence of hemorrhage.
Average correct score by all physi-
cians on all examinations was 77%
(95% CI, 74%-80%) (Table 1). Overall,
of 569 CT scan readings by emergency
physicians, 67% were correct; of 435
readings by neurologists, 83% were
correct; and of 540 readings by radiolo-
gists, 83% were correct. Performance
by neurologists and radiologists was
weakly inversely related to years in
practice. Board certification in emer-
gency medicine did not predict higher
performance. Emergency physicians
who routinely read the cranial CT
scans they ordered scored higher than
those who did not (71% vs 59%; differ-
ence of 12%; 95% CI, 4%-21%).
Percent correct by scan type de-
creased sequentially from easy hemor-
rhages (98%correct) througheasyacute
infarctions, normal scans, normal scans
withcalcification, difficult hemorrhages,
intermediate acute infarctions, normal
scans with old infarctions, and difficult
acuteinfarctions(40%correct) (Table2).
The correlation of percent correct with
scan difficulty (easy, intermediate, or
difficult) substantiatesthevalidityof our
classification schema.
Subjects in the advanced track (those
who responded correctly to the first 5
scans) consistently scored higher on
each type of scan than those in the stan-
dard track (Table 3). For all but those in
the lowest stratum of overall perfor-
mance, easy hemorrhages and easy
acute infarctions were read with near
perfect sensitivity. Scores for other
types of scans improved with increasing
overall skill.
COMMENT
IntheUnitedStates, recombinant tis-
sue plasminogen activator is approved
for use in acute ischemic stroke when
administered within 180 minutes of
symptomonset.
1
During this time inter-
val thepatient must recognizethesymp-
toms, get to a hospital, undergo evalua-
tion, and have a CT scan performed and
interpreted. There will be situations
when the 3-hour limit is rapidly ap-
proaching, no neuroradiologist is avail-
able, and an emergency physician, neu-
rologist, orgeneral radiologist istheonly
physician available to interpret the CT
scan. Ourstudyaskswhetherphysicians
in these specialties are capable, without
additional training, of interpreting the
CTscanwithsufficient sensitivityto de-
Table 1.Performance on First 5 Scans and All Scans by Subject
No. (%) Correct
% of Physicians
Emergency Medicine
(n = 38)
Neurology
(n = 29)
General Radiology
(n = 36) Total
First 5 scans*
0 0 0 3 1
1 (20) 5 0 3 3
2 (40) 26 3 8 14
3 (60) 37 28 22 29
4 (80) 18 48 31 31
5 (100) 13 21 33 22
Average score (SD) 62 (22) 77 (16) 75 (25) 71 (23)
All 15 scans
0-7 (0-46) 8 0 0 3
8-9 (53-60) 34 0 8 15
10-11 (67-73) 34 21 14 24
12-13 (80-87) 24 55 56 43
14-15 (93-100) 0 24 22 15
Average score (SD) 67 (13) 83 (9) 83 (11) 77 (14)
*Scans were 2 difficult hemorrhages, 1 intermediate acute infarction, 1 normal scan, and 1 impostor.
JAMA, April 22/29, 1998Vol 279, No. 16 Cranial CT InterpretationSchriger et al 1295
1998 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ on 02/25/2013
termine if thrombolytics may be admin-
istered safely.
Overall sensitivity for intracerebral
hemorrhage, an absolute contraindica-
tion to thrombolytic therapy, was 82%.
Twolevels of sensitivityfor hemorrhage
may be expected of physicians who in-
terpret cranial CT scans, one based on a
single-case perspective, the other based
on a population perspective. From the
single-case perspective, the administra-
tion of a thrombolytic agent to a patient
with an intracerebral hemorrhage may
be lethal or have other catastrophic con-
sequences. The potential consequences
to the patient and the physician of ad-
ministering thrombolytic therapy after
failing to recognize hemorrhage on the
CT scan mandate that sensitivity for in-
tracerebral hemorrhage be extremely
high, certainly over 95% and ideally
higher than 99%.
A population-based perspective pro-
videsamorerelaxedrequirement forac-
ceptable sensitivity. For instance, in the
624-patient National Instituteof Neuro-
logical Disorders and Stroke study,
there were 10 more survivors in the
treatment group than in the control
group at 3 months.
4
Assuming that 15%
of all acutehemisphericstrokesarehem-
orrhagic,
16
that patients withsmall hem-
orrhagic strokes have prognoses equal
to those with ischemic stroke, and that
all patients with hemorrhagic stroke
who receive thrombolytics die, any sen-
sitivityfor hemorrhage higher than75%
will preserve a potential net death ben-
efit in the treated group (unpublished
data, D.L.S.). By similar reasoning, the
sensitivity for hemorrhage may de-
crease to as low as 22% before negating
improvements in neurologic outcome as
measured with the Barthel Index.
We weigh heavily the single-case per-
spective and believe that, on average,
the sample of physicians inour studydid
not have the skills needed to recognize
hemorrhage on CTscans and determine
which patients may safely receive
thrombolytic therapy. One implication
of this finding is that physicians in these
specialties should not assume that the
standard of care dictates that they
should all be able to make these deci-
sions independently.
Sensitivity for identification of early
signs of major infarction was variable.
Easy acute infarctions were identified
by most physicians, but even the best
performers failed to identify a third of
the difficult acute infarctions. The clini-
cal importance of this finding in patients
withstrokewithinandbeyond3hours of
symptom onset remains to be deter-
mined.
10,17,18
Poorspecificitywoulddepriveeligible
patients the opportunity to benefit from
thrombolytic therapy. While this is less
catastrophic than administering throm-
bolytic agents to patients with contrain-
dications, our study demonstrates that
many physicians will have trouble dif-
ferentiating hemorrhage from calcifi-
cation and acute infarction from old in-
farction.
It is possible that our study results
could represent a biased estimate of na-
tional average performance. Physicians
in our study may not be representative
of the national population of physicians
in each specialty, leading to sampling
bias. Furthermore, physicians taking a
test may achieve results different from
thosetheyachieveinactual practice. Mo-
tivation in the simulated testing situa-
tion could be higher (eg, years of condi-
tioningtoperformwell, desiretowinthe
textbook) or lower (eg, nopatients lifeis
at stake) than in clinical practice. We
cannot predict the direction of this bias.
The inability of participants to view
CT scans on a computer and measure
image density to differentiate calcium
fromblood may have adversely affected
specificity on normal scans with calcifi-
cations. Higher thanideal ambient light-
inginthe testingarea mayhave reduced
overall performance. We intentionally
prevented physicians from stating un-
sure in response to our questions be-
cause we were interested in determin-
Table 2.Percent Correct by Scan Category
Type of Scan
No. of
Scans
Emergency Physicians Neurologists General Radiologists
% Correct
No. of
Readings
Range Correct
by Scan, % % Correct
No. of
Readings
Range Correct
by Scan, % % Correct
No. of
Readings
Range Correct
by Scan, %
Normal 10 83 109 57-100 90 81 77-100 89 98 70-100
Imposter (calcication) 4 77 39 44-100 78 32 43-100 71 49 39-92
Imposter (old infarction) 4 43 42 27-80 59 32 42-100 70 33 20-88
Hemorrhage (easy) 8 94 98 85-100 100 69 100 100 78 100
Hemorrhage (difficult) 10 57 124 25-93 78 99 40-100 80 128 43-93
Acute infarction (easy) 4 82 45 66-100 100 34 100 93 29 80-100
Acute infarction (intermediate) 8 46 97 14-97 77 70 42-100 85 95 33-100
Acute infarction (difficult) 8 13 15 0 44 18 0-100 50 30 0-100
Total 56* 67 569 . . . 83 435 . . . 83 540 . . .
*There were 54 scans in the initial pool, and 2 were added when 2 problem scans were replaced.
Ellipses indicate data not applicable.
Table 3.Percent Correct by Scan Type Stratied by Subjects Overall Performance
n
Easy
Hemorrhage
Difficult
Hemorrhage
Easy Acute
Infarction
Intermediate
Acute Infarction
Difficult Acute
Infarction Normal
Acute Impostor
(Calcication)
Impostor
(Old Infarction)
Advanced Track (N = 21*)
No. of scans read per subject . . . 5 . . . 2 3 2 3
Highest scorers, % 8 . . . 90 . . . 100 66 100 100 100
Lower scorers, % 13 . . . 86 . . . 88 23 96 83 50
Standard Track (N = 82)
No. of scans read per subject 3 3 2 2 . . . 3 2
Higher scorers, % 34 100 81 100 84 . . . 94 77 79
Intermediate scorers, % 32 100 60 98 60 . . . 84 59 38
Lowest scorers, % 16 88 35 59 29 . . . 67 67 29
*Although 23 subjects achieved perfect scores on the initial series, 2 were inadvertently placed in the standard track.
Ellipses indicate scan type not included in test.
Subjects were stratied by overall percent correct into roughly equal groups.
1296 JAMA, April 22/29, 1998Vol 279, No. 16 Cranial CT InterpretationSchriger et al
1998 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ on 02/25/2013
ing what would happen if the physician
were the only person available to
promptly read the scan. This approach
most likely decreased specificity (those
physicians who expressed uncertainty
about a scan often scored it as infarction
orhemorrhagebecausetheydidnot wish
to miss contraindications), but should
not have affected sensitivity.
The few breaks in protocol also could
affect our estimates, but any decrease in
performanceresultingfromthe2ambigu-
ousscanswaslikelybalancedbythe2sub-
jects who were mistakenly placed in the
standard track (where they likely scored
higher than they would have in the ad-
vancedtrack). Giventhemagnitudeof our
results, andthefact that manyphysicians
were able to attain a perfect sensitivity
for hemorrhage, we do not believe that
these testing biases are large enough to
alter our findings substantially.
Our convenience sampling method is
the main threat to the external validity
of the study. While board certification is
an imperfect proxy for skill at interpret-
ingcranial CTscans, all neurologists and
radiologists were board certified, and
the percentage of board-certified emer-
gencyphysiciansinthesampleexceeded
the national percentage of full-time
emergency physicians who are board
certified. Whilewecannot provethat our
sample was representative of the na-
tional population of physicians in these
specialties, we observed that physicians
who acknowledged that they were un-
comfortable reading CT scans often
declined participation, suggesting that
our study most likely did not underesti-
mate physician performance. In addi-
tion, overall physician performance cor-
related with performance on each scan
type, which suggests that the examina-
tion had construct validity.
In conclusion, it appears that while
some members of each of these physi-
cian groups are capable of identifying
hemorrhage with perfect or near-per-
fect sensitivity, the majority of those
tested are not. Board certification in
emergency medicine, neurology, or gen-
eral radiology is an inadequate marker
for such competence. Physicians in-
volved in the care of patients with acute
stroke should ensure that the interpre-
tion of the CTscan reliably identifies in-
tracranial hemorrhage when present.
This may be accomplished by providing
physicians withenhancedtraininginthe
interpretation of cranial CT scans or by
implementing teleradiography or other
systems that facilitate immediate scan
interpretation by qualified readers.
The National Stroke Association (NSA) allowed
us to use their exhibit booth for execution of this
study, and underwrote the cost of the T-shirts and
medical books. The NSA supported these activities
with funds provided by an educational grant from
Knoll Pharmaceutical Company, Whippany, NJ.
Weoffer our sincerethanks tothephysicians who
participated in this study. We also thank WilliamH.
Rogers, PhD, for help with the modeling of sample
sizeconsiderations; Michael Zucker, MD, andEmily
Nicklinfor helpfindingandduplicatingscans; Pablo
Villablanca, MD, neuroradiologist, for aid in scan
interpretations; Danica Barron for data entry and
data checking; and Julea Leshar McGhee for help
with recruitment and enrollment at the study sites.
References
1. Nightingale SL. t-PA approved for acute ische-
mic stroke [From the Food and Drug Administra-
tion]. JAMA. 1996;276:443.
2. Hacke W, Kaste M, Fieschi C, et al. Intravenous
thrombolysis with recombinant tissue plasminogen
activator for acute hemispheric stroke: the Euro-
pean Cooperative Acute Stroke Study (ECASS).
JAMA. 1995;274:1017-1025.
3. Multicenter Acute Stroke TrialItaly (MAST-I)
Group. Randomised controlled trial of streptokinase,
aspirin, andcombinationof bothintreatment of acute
ischaemic stroke. Lancet. 1995;346:1509-1514.
4. National Institute of Neurological Disorders and
Strokert-PAStrokeStudyGroup. Tissueplasmino-
gen activator for acute ischemic stroke. N Engl J
Med. 1995;333:1581-1587.
5. Donnan GA, Davis SM, Chambers BR, et al.
Streptokinase for acute ischemic stroke with rela-
tionshipto time of administration. JAMA. 1996;276:
961-966.
6. Multicenter Acute Stroke TrialEurope Study
Group. Thrombolytic therapy with streptokinase in
acute ischemic stroke. N Engl J Med. 1996;335:
145-150.
7. Special Writing Committee of the Stroke Coun-
cil, American Heart Association. Guidelines for
thrombolytic therapyfor acute stroke. Stroke. 1996;
27:1711-1718.
8. Quality Standards Subcommittee of the Ameri-
can Academy of Neurology. Practice advisory:
thrombolytic therapy for acute ischemic stroke:
summary statement. Neurology. 1996;47:835-839.
9. National Stroke Association consensus state-
ment: stroke: the first hours: emergency evaluation
and treatment guidelines. In: Stroke Clinical Up-
dates. Englewood, Colo: National Stroke Associa-
tion; 1997.
10. National Institute of Neurological Disorders
and Stroke rt-PA Stroke Study Group. Intracere-
bral hemorrhageafterintravenoust-PAtherapyfor
ischemic stroke. Stroke. 1997;28:2109-2118.
11. Moulin T, Besson G, Crepin-Leblond T, et al.
HemorrhagictransformationsinMAST-Etrial: pre-
dictive factors [abstract]. Cerebrovascular Dis.
1996;6:182.
12. Larrue V, von Kummar R, del Zoppo G, Bluhm-
ki E. Hemorrhagic transformationinacuteischemic
stroke: potential contributing factors in the Eu-
ropean Cooperative Acute Stroke Study. Stroke.
1997;28:957-60.
13. Ebel RL, Frisbie DA. Essentials of Educa-
tional Measurement. 4th ed. Englewood Cliffs, NJ:
Prentice-Hall International Inc; 1986:335-339.
14. Wainer H, Dorans NJ, Flangher R, et al. Com-
puterized Adaptive Testing: A Primer. Hillsdale,
NJ: Lawrence Erlbaum Associates Inc Publishers;
1990.
15. Stata Corporation. Stata 5.0 Statistical Soft-
wareUsersGuide. CollegeStation, Tex: StataCorp;
1997:235-239.
16. SaccoR. Classificationof stroke. In: Fisher M, ed.
Clinical Atlas of Cerebrovascular Disorders. Lon-
don, England: MosbyYear Book Inc; 1994:2.1-2.23.
17. von Kummer R, Allen KL, Holle R, et al. Acute
stroke: usefulness of early CT findings before
thrombolytic therapy. Radiology. 1997;205:327-
333.
18. National Institute of Neurological Disorders
and Stroke rt-PA Stroke Study Group. General-
ized efficacy t-PA for acute stroke: subgroup
analysis of the NINDS stroke trial. Stroke. 1997;
28:2119-2125.
JAMA, April 22/29, 1998Vol 279, No. 16 Cranial CT InterpretationSchriger et al 1297
1998 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ on 02/25/2013