Downloaded from http://pmj.bmj.com/ on June 14, 2017 - Published by group.bmj.
com
PGMJ Online First, published on December 23, 2016 as 10.1136/postgradmedj-2016-134491
1
National Trauma Research
Institute, Monash Alfred Injury
Network, Melbourne, Australia
2
Monash University,
Melbourne, Australia
3
Trauma Service, The Alfred
Hospital, Melbourne, Australia
4
Department of Epidemiology
& Preventive Medicine,
Monash University, Clayton,
Victoria, Australia
5
Emergency & Trauma Centre,
The Alfred Hospital,
Melbourne, Australia
6
Department of Radiology, The
Alfred Hospital, Melbourne,
Australia
Correspondence to
Lachlan Evans, National
Trauma Research Institute,
85-89 Commercial Road,
Melbourne,
VIC 3004, Australia;
lreva3@student.monash.edu
Received 30 August 2016
Revised 26 October 2016
Accepted 3 December 2016
To cite: Evans LR,
Fitzgerald MC, Mitra B,
et al. Postgrad Med J
Published Online First:
[please include Day Month
Year] doi:10.1136/
postgradmedj-2016-134491
Copyright Article author (or their employer) 2016. Produced by BMJ Publishing Group Ltd under licence.
Original article
Emergency department interpretation of CT of the
brain: a systematic review
Lachlan R Evans,1,2 Mark C Fitzgerald,1,3 Biswadev Mitra,1,4,5 Dinesh Varma6
ABSTRACT
Background and objectives CT of the brain (CTB) is
one of the most common radiological investigations
performed in the emergency department (ED).
Emergency clinicians rely upon this imaging modality to
aid diagnosis and guide management. However, their
capacity to accurately interpret CTB is unclear. This
systematic review aims to determine this capacity and
identify the potential need for interventions directed
towards improving the ability of emergency clinicians in
this important area.
Methods A systematic review of the literature was
conducted without date restrictions. We searched
MEDLINE, EMBASE and Cochrane databases and studies
reporting the primary outcome of concordance of CTB
interpretation between a non-radiologist and a radiology
specialist were identied. Studies were assessed for
heterogeneity and a subgroup analysis of pooled data
based on medical specialty was carried out to specically
identify the concordance of ED clinicians. The quality of
evidence was assessed using the GRADE criteria.
Results There were 21 studies included in this review.
Among the included studies, 12 reported on the
concordance of emergency clinicians, 5 reported on
radiology trainees and 4 on surgeons. Clinical and
statistical heterogeneity between studies was high
(I2=97.8%, p<0.01). The concordance in the emergency
subgroup was the lowest among all subgroups with a
range of 0.630.95 and a clinically signicant error rate
ranging from 0.02 to 0.24.
Conclusions Heterogeneity and the presence of bias
limit our condence in these ndings. However, the
variance in the interpretation of CTB between emergency
clinicians and radiologists suggests that interventions
towards improving accuracy may be useful.
INTRODUCTION
CT of the brain (CTB) is among the most common
investigations ordered in the emergency depart-
ment (ED). In total, 1 in 14 patients who present
to the ED are investigated by this imaging modality
and this increases to 1 in 7 for the population aged
>65 years.1 Notably, there has been an upward
trend in the use of CTB over the last two decades
and an expansion of the clinical scenarios in which
it plays a crucial diagnostic and therapeutic role.2
The centrality of CTB to the diagnosis of neuro-
logical disease in the ED is emphasised by the
Australian College for Emergency Medicines
(ACEMs) curriculum requiring graduating trainees
to be able to interpret CTB with an expert level
of prociency.3 While most tertiary hospitals have a
system of formal radiology oversight for emergency
medical staff, the time-critical nature of many acute
neurological emergencies mandates the need for
Evans LR, et al. Postgrad Med J 2016;0:16. doi:10.1136/postgradmedj-2016-134491
rapid and accurate ED interpretation of CTB.
Recent Australian data that demonstrate increases
of up to 17% in the number of out-of-hours
CTBs ordered by ED staff compound this problem
as out-of-hours reports are usually provided by
radiology trainees or are unavailable until the next
morning.4
Traumatic brain injury (TBI) represents approxi-
mately half of all clinical indications for CTB in the
ED.1 Patient outcomes in this cohort are strongly
linked to the rapid identication and treatment of
intracranial pathology.5 The management of path-
ologies such as expanding intracranial haemorrhage
and raised intracranial pressure requires swift
neurosurgical and medical intervention, which are
facilitated by immediate CTB interpretation in the
ED.6 Advanced trauma systems have streamlined
this process; however, it is frequently noted that
the period from CT scanning to the initiation of
appropriate management is unsatisfactory.7 While
the factors underpinning this problem are unclear,
the ability of emergency clinicians to assess the
images and rapidly act upon their ndings is a vital
component of enhancing the care of patients
with TBI.
Despite this there is a paucity of evidence that
describes the capacity of emergency medical staff to
accurately interpret CTB. This systematic review
aims to determine this capacity and outline the
need, if any, for further interventions towards
improving this fundamental skill.
METHODS
This systematic review was conducted in accord-
ance with the preferred reporting items for system-
atic reviews and meta-analysis guidelines.8 The
methodology of this review was registered with the
PROSPERO online record of systematic reviews on
11 April 2016.9
Search strategy
MEDLINE, EMBASE and Cochrane Library were
searched in April 2016 using a broad strategy that
included all studies that reported on CTB interpret-
ation by non-radiologists. The search strategy used
a combination of keywords and text words that
were matched to MeSH and centred on the con-
cepts of brain injury, computed tomography and
clinical competence. There were no date or
methodology restrictions. Limits included English-
language articles with non-paediatric populations
only. Furthermore, sources of grey literature such as
Google Scholar as well as the reference lists of all
eligible studies were searched for additional
articles.
1
 Downloaded from http://pmj.bmj.com/ on June 14, 2017 - Published by group.bmj.com

Original article

study characteristics such as the sample size and the authors

denition of clinically signicant error.

Analysis
We considered the quality of the evidence for each outcome
according to the GRADE criteria, which generates a rating of the
evidence by evaluating the study design and a number of modify-
ing factors that can upgrade or (more likely) downgrade the
quality of evidence.10 The concordance data from each study
were pooled using the DerSimonian-Laird random effects model
and the pooled effect size was reported as a proportion with 95%
CI. Heterogeneity was assessed using the I2 statistic. A subgroup
analysis was planned in order to identify the population of inter-
est, emergency medical staff. Statistical analysis was conducted
using Stata V.11.0 (StataCorp, College Station, Texas, USA).

Figure 1 Description of study selection using the preferred reporting
items for systematic reviews and meta-analysis method.
Study selection
The authors reviewed abstracts and full-text articles where rele-
vant, with any inconsistency resolved by consensus. Eligible
studies reported a primary outcome that measured the concord-
ance of non-contrast CTB interpretation between non-radiology
medical specialists and qualied radiologists. Exclusion criteria
included the use of paediatric patient populations, a gold stand-
ard for CTB interpretation non-equivalent to a radiology special-
ist and unclear or insufcient reporting of the primary outcome.
Data extraction
Data from each eligible study were collected using a standard
template that focused on methodology, results and relevant
RESULTS
There were 41 eligible studies identied for full-text review,
after which a further 20 were excluded. Figure 1 depicts the
process of study selection.
Among the 20 studies that were excluded on full-text review,
17 did not meet eligibility criteria and the remainder met exclu-
sion criteria. Of the 13 studies that did not report a relevant
primary outcome, 5 assessed diagnostic variance on CTB
between radiology specialists, 5 compared different CT tech-
nologies and the nal 3 studies only reported the incidence of
CTB use. All four studies that failed to report on the interpret-
ation of CTB assessed cerebral angiography instead. There were
two further studies that were excluded due to poor reporting of
the primary outcome. The rst did not clearly describe the rele-
vant outcome and the second reported clinician performance on
a non-clinical test from which the authors were unable to
extract concordance data. The nal study excluded was an
abstract only.
Table 1 reports the characteristics of the 21 studies identied
as appropriate for quantitative synthesis. There was signicant

Table 1 Characteristics of studies selected in the systematic review

Author (year) Country Participants Methodology Total number of scans Number of abnormal scans (%)

Al-Reesi11 (2010) Oman ED clinicians Retrospective 442 82 (18.6%)

Alfaro12 (1994) USA ED clinicians Prospective 555 256 (46.1%)
Ardic13 (2015) Turkey ED clinicians Prospective 525 94 (17.9%)
Arendts14 (2003) Australia ED clinicians Prospective 1282 551 (43.1%)
Boyle15 (2009) UK ED clinicians Cross-sectional 243 81 (33.3%)
Cheung16 (2015) Hong Kong ED clinicians Retrospective 1716 283 (16.5%)
Dolatabadi17 (2013) Iran ED clinicians Prospective 544 259 (47.6%)
Harding18 (2010) UK ED clinicians Prospective 547
Khan19 (2013) Saudi Arabia ED clinicians Prospective 241
Khoo20 (2007) Australia ED clinicians Prospective 315 84 (28%)
Mucci21 (2005) UK ED clinicians Retrospective 100 41 (41%)
Talebian22 (2015) Iran ED clinicians Prospective 450 194 (43.1%)
Chun23 (2010) USA NS Retrospective 50
Dourado24 (2015) Brazil NS Prospective 227 164 (72.2%)
Mukerji25 (2009) UK NS Prospective 192 88 (30.1%)
Vorhies26 (2002) USA General surgery trainees Prospective 47 10 (21.3%)
Erly27 (2002) USA RT Prospective 1324 554 (42%)
Miyakoshi28 (2009) Japan RT Retrospective 6852
Roszler29 (1991) USA RT Prospective 289 64 (22%)
Strub30 (2006) USA RT Prospective 5206
Wysoki31 (1998) USA RT Prospective 419
ED, emergency department; NS, neurosurgeons; RT, radiology trainees.

2 Evans LR, et al. Postgrad Med J 2016;0:16. doi:10.1136/postgradmedj-2016-134491

 Downloaded from http://pmj.bmj.com/ on June 14, 2017 - Published by group.bmj.com

Original article

clinical heterogeneity between these studies in terms of the def- DISCUSSION

inition of clinically signicant misinterpretation (CSM), the stat-
istical measure of accuracy and the way in which the clinicians
accuracy was assessed. Due to this heterogeneity and the need
to identify the performance of emergency clinicians specically,
studies were separated into subgroups based on the medical spe-
cialty of participants and were analysed as such.
The largest subgroup reported the accuracy of emergency
medical staff, and of these 12 studies, 11 reported the concord-
ance between ED interpretation and a radiology specialist.
Boyle15 reported the performance of emergency physicians on a
non-clinical test that scored the CTBs in relation to their sub-
jective difculty and could not be included in quantitative ana-
lysis. Table 2 describes the ndings of each study and the
measures of accuracy that were used.
Upon visual inspection of the forest plot (gure 2), it was
clear that signicant heterogeneity existed between the 11 ED
studies. This was conrmed formally through the calculation of
an I2 statistic of 97.8% ( p<0.01). The same analysis was con-
ducted using data on the proportion of CSMs, and this was
similarly heterogeneous (gure 3). A pooled effect size of ED
concordance of 0.82 (95% CI 0.78 to 0.85) with a range of
0.630.95 was noted. The pooled proportion of CSMs was
0.07 (95% CI 0.04 to 0.09) with a range of 0.020.24. Pooled
analysis of the neurosurgery and general surgery subgroups was
not conducted due to inconsistent reporting of concordance.
The remaining subgroup contained ve studies that reported the
accuracy of radiology trainees. The pooled effect size was not
calculated; however, the range of concordance was 0.910.97
with a CSM rate of 0.0010.025.
The GRADE criteria10 were used to appraise the strength of
the evidence, with each selected study being observational in
design and starting as low-level evidence. Due to the likelihood
of signicant bias and the inconsistency of results, this was
downgraded further and hence the quality of evidence available
on this topic was classied as very low. Table 3 describes the
process of GRADE criteria10 analysis.
Concordance of emergency medical staff with the gold standard
was low and ranged from 0.63 to 0.95. The rate of CSMs
across studies was also substantial and ranged from 0.02 to
0.24. Statistical and clinical heterogeneity among studies, in
combination with the low level of evidence imparted by study
designs, limit our condence in these ndings.
All but one study in the ED subgroup extracted data from the
patients medical record, with Mucci et al21 instead selecting
100 CTBs from the hospital radiology system and testing the
participants in a non-clinical environment. Among the remain-
ing 10 studies, only Al-Reesi et al11 and Cheung et al16 con-
ducted retrospective reviews and found the concordance
between ED and the nal radiology report to be 0.93 (95% CI
0.91 to 0.95) and 0.91 (95% CI 0.90 to 0.92), respectively.
The eight studies that obtained data prospectively had signi-
cantly heterogeneous methods in regards to whether the ED
interpretation was documented freehand or on a standard tem-
plate, the blinding of participants to the clinical indication for
the scan and the level of training of the participants. Of the 11
ED studies, 4 did not delineate between qualied emergency
physicians and emergency trainees while 4 included emergency
physicians only and a single study exclusively assessed trainees.
Arendts14 and Dolatabadi17 assessed overall accuracy as well as
comparing emergency physicians to trainees as a secondary
outcome and found no difference in accuracy between the
groups. Notably, Ardic13 conducted the one study that included
emergency trainees only and found the accuracy to be 0.95
(95% CI 0.93 to 0.97), which is the highest level of concord-
ance among studies selected for this systematic review.
In translating these ndings to the clinical environment, the
most pertinent data relate to the proportion of CSMs. While
Alfaro12 reported a high CSM rate of 0.24 (95% CI 0.20 to
0.27), this was the oldest study included in the review and the
authors also noted that only 3 of the 133 CSMs resulted in
documented patient harm. If this study is excluded, the range of
CSMs in the ED is 0.020.11. Nevertheless, the rate of CSM

Table 2 Results of studies selected in the systematic review

Author (year) Concordance Discordance CSM rate in total Sensitivity (95% CI) Specificity (95% CI) Kappa coefficient

Al-Reesi11 (2010) 0.93 0.07 0.05 0.78 (0.670.86) 0.96 (0.930.98) 0.83
Alfaro12 (1994) 0.63 0.37 0.24
Ardic13 (2015) 0.95 0.05 0.03 0.93 0.93 0.82
Arendts14 (2003) 0.85 0.15 0.06 0.69
Boyle15 (2009) 0.92 (0.870.96) 0.75 (0.650.84)
Cheung16 (2015) 0.91 0.09 0.09 0.66 (0.600.71) 0.96 (0.950.97)
Dolatabadi17 (2013) 0.84 0.16
Harding18 (2010) 0.64 0.36 0.02
Khan19 (2013) 0.87 0.13 0.06 0.96 (0.930.99) 0.63 (0.460.77) 0.64
Khoo20 (2007) 0.67 0.33 0.11 0.57 (0.450.69) 0.63 (0.460.77) 0.24
Mucci21 (2005) 0.87 0.13 0.04
Talebian22 (2015) 0.86 0.14 0.03 0.62 0.86 0.68
Chun23 (2010) 0.92 0.08 0.79 0.88 0.57
Dourado24 (2015) 0.76
Mukerji25 (2009) 0.70
Vorhies26 (2002) 0.94 0.06 0.02 0.89 1.00
Erly27 (2002) 0.91 0.09 0.02
Miyakoshi28 (2009) 0.97 0.03 0.01
Roszler29 (1991) 0.96 0.04 0.02
Strub30 (2006) 0.96 0.04 0.001
Wysoki31 (1998) 0.96 0.04 0.02
CSM, clinically significant misinterpretation.

Evans LR, et al. Postgrad Med J 2016;0:16. doi:10.1136/postgradmedj-2016-134491 3

 Downloaded from http://pmj.bmj.com/ on June 14, 2017 - Published by group.bmj.com

Original article

Figure 2 Forest plot describing the rates of concordance in the emergency department subgroup.

Figure 3 Forest plot describing the rates of clinically signicant misinterpretations in the emergency department subgroup.

for CTB is signicantly higher than that reported for other
imaging modalities interpreted by the ED staff. Nitowski32
reports a CSM rate of 0.002 for emergency physicians interpret-
ing plain radiography of the limbs, while Safari et al33 found
a total discrepancy rate of 0.015 for interpretation of the
chest X-ray.
The primary comparison made in this analysis was between
the ED and radiology trainee subgroups. First, the range of con-
cordance was wider in the ED subgroup and reects the relative
4 Evans LR, et al. Postgrad Med J 2016;0:16. doi:10.1136/postgradmedj-2016-134491
homogeneity of studies that assessed radiology trainees. As
expected, the concordance data showed a higher level of accur-
acy in the radiology trainee subgroup compared with the ED
staff. Furthermore, the rate of CSMs for radiology trainees
(0.0010.025) is consistent with the approximate CT misinter-
pretation rate of 0.02 that is reported in the radiology litera-
ture.28 Gallagher et al,34 who were excluded from quantitative
analysis, directly compared the performance of radiologists,
radiology trainees, neuroradiographers and emergency clinicians
 Downloaded from http://pmj.bmj.com/ on June 14, 2017 - Published by group.bmj.com

Original article

Table 3 Summary of GRADE criteria10 analysis

Number of studies Limitations Inconsistency Indirectness Imprecision Publication bias Quality

Outcome: accuracy of ED interpretation

11 Observ. (low) Serious risk of bias (1) Serious (1) None (0) None (0) None likely given value of both Very low (2)
negative and positive outcomes (0)
No upgrading modifiers
Outcome: accuracy of RT interpretation
5 Observ. (low) Serious risk of bias (1) None (0) None (0) None (0) None likely given value of both Very low (1)
negative and positive outcomes (0)
No upgrading modifiers
ED, emergency department; Observ. observational study; RT, radiology trainee.

and found that the latter had the highest error rate for CTB light of the frequency of CTBs being requested these ndings
interpretation. As noted previously, the ACEM concept of indicate the importance of educational and institution-based
expert prociency correlates practically with the level of skill interventions directed towards improving the accuracy of CTB
expected of a consultant in that eld. Therefore, comparison of interpretation by emergency clinicians.
ED performance to both radiology specialists and trainees indi-
cates a possible deciency in the skillset required of specialist
emergency physicians.
The studies that reported on neurosurgical interpretation of Main messages
CTB varied between prospective evaluation of images and retro-
spective, exam-based assessment. Chun23 reported a concord-
CT of the brain is a common and important emergency
ance rate of 0.92 (95% CI 0.90 to 0.95). However, this study
department investigation.
was limited by the inclusion of only ve participants and by
The capacity of emergency clinicians to accurately interpret
focusing on scoring systems that are not commonly used in clin-
this type of imaging is unclear.
ical practice. All three studies in the neurosurgery subgroup
We identied 21 studies describing the concordance of
reported a moderatehigh level of inter-rater agreement with
radiologists with non-radiologists; 12 reported on emergency
Cohens coefcients of 0.570.76. The nal study by Vorhies
clinicians, 5 on radiology trainees and 4 on surgeons.
et al26 reported the accuracy of senior general surgical trainees
The rate of concordance was lowest in the emergency
interpreting a variety of modalities and noted a CTB accuracy of
clinician subgroup with a range of 0.630.95 and a clinically
0.94 (95% CI 0.90 to 0.97). This study was restricted in regards
signicant error rate of 0.020.24.
to the small number of scans included (n=47) and the exclusion
Given the frequency with which CT of the brain is used in
of any scan interpreted in the presence of a senior surgeon.
the emergency department, interventions towards improving
The major limitation to this review is the presence of signi-
accuracy may be warranted.
cant heterogeneity. As described previously, the variability in
study design and the lack of a consistent method of assessing
accuracy have impeded meta-analysis. Substantial bias was
present in each individual study, most commonly in regards to
Current research questions
the baseline characteristics of selected participants and their
level of experience. The total number of scans interpreted in
each study varied considerably and the proportion of those 1. The use of CT of the brain in the emergency department has
scans that were abnormal had a similarly wide range (0.16 decreased with the development of MRI:
0.72). The majority of studies noted that there was a likely rela- a. True
tionship between the proportion of abnormal scans and the b. False
number of misinterpretations. There was further ambiguity in 2. Suspected traumatic brain injury is the most common
the way in which a participants interpretation was classied as indication for CT of the brain in the emergency department:
concordant or discordant. Several studies noted that incomplete a. True
or poor documentation frequently impaired the extraction of b. False
data, which was resolved by the investigators subjectively decid- 3. Emergency clinicians and radiology trainees have been
ing the result. The variety of study limitations and the signi- shown to interpret CT of the brain with similar rates of
cant presence of bias is reected in the formal classication of accuracy:
the quality of evidence on this topic as very low, as dened by a. True
the GRADE criteria.10 b. False
4. The error rate for emergency clinicians interpreting CT of the
brain is similar to plain radiography of the chest:
CONCLUSION
a. True
Statistical and clinical heterogeneity limit the strength of this
b. False
reviews ndings. However, the individual conclusions drawn
5. Ongoing educational interventions are indicated to reduce
from each study indicate variance between ED interpretation of
the error rate for emergency clinicians interpreting CT of the
CTB and the nal radiology report. Of note, the rate of clinic-
brain:
ally signicant errors ranged from 0.02 to 0.24. High-quality
a. True
prospective research is required to further elucidate the ability
b. False
of emergency clinicians in this important area. Nonetheless, in
Evans LR, et al. Postgrad Med J 2016;0:16. doi:10.1136/postgradmedj-2016-134491 5
 Downloaded from http://pmj.bmj.com/ on June 14, 2017 - Published by group.bmj.com
Original article
Acknowledgements The authors acknowledge the authors of all original studies
included in this review.
Contributors LRE was responsible for the planning and execution of the systematic
review, drafting of the article and submission to the journal. BM reviewed the search
strategy and carried out the statistical analyses. BM, MCF and DV contributed to the
editing and revision of the manuscript. As rst author and guarantor, LE is
ultimately responsible for the content and format of the nal manuscript.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
REFERENCES
1 Raja AS, Andruchow J, Zane R, et al. Use of neuroimaging in US emergency
departments. Arch Intern Med 2011;171:2602.
2 Larson DB, Johnson LW, Schnell BM, et al. National trends in CT use in the
emergency department: 1995-2007. Radiology 2011;258:16473.
3 Australian College for Emergency Medicine. ACEM Curriculum Framework 2015.
p. 68.
4 Goergen SK, Grimm J, Paul E, et al. Audit of demand for after-hours CT scanning
services in RANZCR-accredited training departments. J Med Imaging Radiat Oncol
2016;60:3546.
5 Hrtl R, Gerber LM, Iacono L, et al. Direct transport within an organized state
trauma system reduces mortality in patients with severe traumatic brain injury.
J Trauma 2006;60:12506; discussion 6.
6 Seelig JM, Becker DP, Miller JD, et al. Traumatic acute subdural hematoma: major
mortality reduction in comatose patients treated within four hours. N Engl J Med
1981;304:151118.
7 Mellanby E, Clark R, McLellan S. Time from CT scan to surgery in traumatic brain
injury. Room for improvement? Anaesthesia. Conference:Scottish Society of
Anaesthetists Annual Meeting 2011 Crieff, Perthshire United Kingdom. Conference
Publication: 66.
8 Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews
and meta-analyses: the PRISMA statement. J Clin Epidemiol 2009;62:100612.
9 Evans L, Fitzgerald M, Mitra B. Assessing the accuracy of non-radiologists in
interpreting computed tomography of the brain. PROSPERO 2016:
CRD42016037210. http://www.crd.york.ac.uk/PROSPERO/display_record.asp?
ID=CRD42016037210
10 Guyatt G, Oxman AD, Akl EA, et al. GRADE guidelines: 1. Introduction-GRADE
evidence proles and summary of ndings tables. J Clin Epidemiol
2011;64:38394.
11 Al-Reesi A, Stiell IG, Al-Zadjali N, et al. Comparison of CT head interpretation
between emergency physicians and neuroradiologists. Euro J Emerg Med
2010;17:2802.
12 Alfaro D, Levitt MA, English DK, et al. Accuracy of interpretation of cranial
computed tomography scans in an emergency medicine residency program. Ann
Emerg Med 1994;25:16974.
13 Ardic S, Guneysel O. Emergency medicine residents can assess cranial computed
tomography scans consistently with radiologists. J Acute Med 2015;5:925.
14 Arendts G, Manovel A, Chai A. Cranial CT interpretation by senior emergency
department staff. J Med Imaging Radiat Oncol 2003;47:36874.
6 Evans LR, et al. Postgrad Med J 2016;0:16. doi:10.1136/postgradmedj-2016-134491
15 Boyle A, Staniciu D, Lewis S, et al. Can middle grade and consultant emergency
physicians accurately interpret computed tomography scans performed for head
trauma? Cross-sectional study. Emerg Med J 2009;26:5835.
16 Cheung LW, Shih YN, Leung CS, et al. How accurate are emergency department
medical staff in the interpretation of head injury related computed tomography?
Hong Kong J Emerg Med 2015;22:2417.
17 Arhami Dolatabadi A, Baratloo A, Rouhipour A, et al. Interpretation of computed
tomography of the head: Emergency physicians versus radiologists. Trauma Mon
2013;18:869.
18 Harding J, Craig M, Jakeman N, et al. Emergency physician interpretation of head
CT in trauma and suspected subarachnoid haemorrhageis it viable? An audit of
current practice. Emerg Med J 2010;27:11620.
19 Khan A, Qashqari S, Al-Ali AA. Accuracy of non-contrast CT brain interpretation by
emergency physicians: a cohort study. Pak J Med Sci 2013;29:54953.
20 Khoo NC, Duffy M. Out of hours non-contrast head CT scan interpretation by
senior emergency department medical staff. Emerg Med Australas 2007;19:1228.
21 Mucci B, Brett C, Huntley LS, et al. Cranial computed tomography in trauma: The
accuracy of interpretation by staff in the emergency department. Emerg Med J
2005;22:53840.
22 Talebian MT, Kavandi E, Farahmand S, et al. Comparing the brain CT scan
interpretation of emergency medicine team with radiologists report and its impact
on patients outcome. Emerg Radiol 2015;22:2618.
23 Chun KA, Manley GT, Stiver SI, et al. Interobserver variability in the
assessment of CT imaging features of traumatic brain injury. J Neurotrauma
2010;27:32530.
24 Dourado JC, Pereira JL, Albuquerque LA, et al. Indices of agreement between
neurosurgeons and a radiologist in interpreting tomography scans in an emergency
department. Arq Neuropsiquiatr 2015;73:68891.
25 Mukerji N, Cahill J, Paluzzi A, et al. Emergency head CT scans: can neurosurgical
registrars be relied upon to interpret them? Br J Neurosurg 2009;23:15861.
26 Vorhies RW, Harrison PB, Smith RS, et al. Senior surgical residents can accurately
interpret trauma radiographs. Am Surg 2002;68:2215; discussion 5-6.
27 Erly WK, Berger WG, Krupinski E, et al. Radiology resident evaluation of
head CT scan orders in the emergency department. AJNR Am J Neuroradiol
2002;23:1037.
28 Miyakoshi A, Nguyen QT, Cohen WA, et al. Accuracy of preliminary interpretation of
neurologic CT examinations by on-call radiology residents and assessment of patient
outcomes at a level I trauma center. J Am Coll Radiol 2009;6:86470.
29 Roszler MH, McCarroll KA, Rashid T, et al. Resident interpretation of emergency
computed tomographic scans. Invest Radiol 1991;26:3746.
30 Strub WM, Vagal AA, Tomsick T, et al. Overnight resident preliminary interpretations
on CT examinations: should the process continue? Emerg Radiol 2006;13:1923.
31 Wysoki MG, Nassar CJ, Koenigsberg RA, et al. Head trauma: CT scan interpretation
by radiology residents versus staff radiologists. Radiology 1998;208:1258.
32 Nitowski LA, OConnor RE, Reese CL IV. The rate of clinically signicant plain
radiograph misinterpretation by faculty in an emergency medicine residency
program. Acad Emerg Med 1996;3:7829.
33 Safari S, Baratloo A, Negida AS, et al. Comparing the interpretation of traumatic chest
x-ray by emergency medicine specialists and radiologists. Arch Trauma Res 2014;3.
34 Gallagher FA, Tay KY, Vowler SL, et al. Comparing the accuracy of initial head CT
reporting by radiologists, radiology trainees, neuroradiographers and emergency
doctors. Br J Radiol 2011;84:10405.
 Downloaded from http://pmj.bmj.com/ on June 14, 2017 - Published by group.bmj.com

Emergency department interpretation of CT

of the brain: a systematic review
Lachlan R Evans, Mark C Fitzgerald, Biswadev Mitra and Dinesh Varma

Postgrad Med J published online December 23, 2016

Updated information and services can be found at:

http://pmj.bmj.com/content/early/2016/12/23/postgradmedj-2016-134
491

These include:

References This article cites 30 articles, 5 of which you can access for free at:
http://pmj.bmj.com/content/early/2016/12/23/postgradmedj-2016-134
491#BIBL
Email alerting Receive free email alerts when new articles cite this article. Sign up in the
service box at the top right corner of the online article.

Notes

To request permissions go to:

http://group.bmj.com/group/rights-licensing/permissions

To order reprints go to:

http://journals.bmj.com/cgi/reprintform

To subscribe to BMJ go to:

http://group.bmj.com/subscribe/